(Revised) This proposal is designed to examine the role of alcohol consumption and exposure to risky settings in the victimization of young women. The proposal is in response to NIAAA PA 91-75 and builds upon the PI's Scientist Development Award (K21 AA00186), which explores the relationship between alcohol and risk-taking among women. Although victimization and HIV infection have similar antecedents, they rarely have been examined together. A conceptual model is proposed and tested. This model suggests that childhood abuse and risk- prone personality (high sensation- seeking, high negative affect, low assertiveness) lead women to engage in risky behaviors in adulthood (heavy alcohol and drug use and exposure to risky settings such as bars). These risky behaviors, in turn, increase the likelihood of experiencing both victimization and HIV infection. Women who experience victimization may subsequently engage in heavier drinking and higher levels of risky behavior, thereby increasing their chances of revictimization as well as subsequent HIV infection. The model will be tested with a three-wave, cross- lagged panel design, using a representative sample of 1000 unmarried women, ages 18-30, recruited through random digit dialing. Information on 1) victimization, 2) personality, 3) substance use, 4) behavior, and 5) exposure to risky settings will be collected through questionnaires administered at 1-year intervals. These longitudinal data will be analyzed using structural equation modeling. Findings will provide valuable information on the impact of alcohol use and other risky behaviors on victimization, on the relationship between victimization and HIV- risk, and on the link between victimization and revictimization. In addition to examining the interrelationships among risky behaviors and negative outcomes across time, we also will focus on specific incidents of aggression. Women who report experiences of victimization prior to entry into the study will participate in a structured interview designed to assess the role of alcohol consumption and other contextual variables in specific aggressive incidents involving either physical force or verbal coercion. These findings will supplement the findings of the panel study with information on the proximal effects of alcohol consumption on perceptions and behaviors relevant to victimization.